INDEX

Explanations

instances where someone expresses positive feelings about something

oai_token-act-pair · gpt-3.5-turbo

sentiments related to feeling "good" or "bad" about various experiences or situations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DragonMagazine

-0.80

hiba

-0.79

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

-0.77

hesis

-0.75

Write

-0.73

hai

-0.72

ulum

-0.71

SPONSORED

-0.70

Written

-0.70

------------------------

-0.70

POSITIVE LOGITS

 reforming

0.80

 donating

0.79

 marrying

0.79

 reducing

0.79

 quitting

0.78

 preserving

0.77

 committing

0.76

 trusting

0.76

 losing

0.75

how

0.75

Activations Density 0.063%

instances where someone expresses positive feelings about something

sentiments related to feeling "good" or "bad" about various experiences or situations

No Comments

No Known Activations